F0 Lowering Research Articles

The Use of Prosody in Disambiguation in Mandarin

Abstract This article examined the clarification of surface structure ambiguity via prosody in Mandarin. An acoustical analysis and a perceptual study were carried out for this purpose: the acoustical analysis was aimed at identifying the prosodic cues for disambiguation, and the perceptual study explored how syntactical structure was recovered via prosody. The acoustical analysis of syntactically ambiguous pairs resulting from different bracketings revealed that prosodic disambiguation consisted of pause insertion, F₀ lowering, intensity reduction, and preboundary lengthening. Moreover, laryngealization frequently was observed at syntactic breaks. The perceptual study of ambiguous pairs of edited speech tokens indicated that (1) pause was a more powerful cue than preboundary lengthening in syntactical recovery, and (2) only a large amount of preboundary lengthening led to successful parsing of ambiguous sentences.

An MRI study of the physiological mechanism of F0 lowering

The phenomena of F0 lowering in utterances, widely observed in a word accent or near a phrase boundary, is not sufficiently explained by the relaxation of F0 raising musculature, and the physiological mechanism for F0 lowering has been a question to be examined. Morphological observation of laryngeal framework was performed in this study using the magnetic resonance imaging (MRI) during phonation. Midsagittal images near the larynx were taken in different F0 levels in four male speakers. Relative positional changes among laryngeal components were measured to explore the chain of mechanisms that allows vocal fold shortening in low F0. In the results, vertical movements of the larynx were consistently observed to be correlated with F0 levels. Movements of the hyoid bone and changes in laryngeal cavity size were also found. A rotation of the cricoid cartilage was always associated with laryngeal descent towards lower F0. This rotation occurred to the direction of shortening the vocal folds, and its mechanism was confirmed by vertical sliding motion of the posterior plate of the cricoid cartilage along the physiological curvature of the cervical vertebrae. This mechanism can account for the well-known tendencies of laryngeal descent and strap muscle activity in F0 lowering.

Perceptual shift of accent in English

It has been found that the perception of an accentual high tone (H) in Japanese is determined by both the F0 peak location and the F0 drop rate thereafter. The actual F0 peak need not occur on the accented syllable: If the peak is approximately within the first half of the postaccent syllable, and the drop rate is greater than a certain value, the H will be associated with the accented syllable [Y. Hasegawa and K. Hata, J. Acoust. Soc. Am. Suppl. 1 83, S29 (1988); K. Hata and Y. Hasegawa, J. Acoust. Soc. Am. Suppl. 1 84, S156 (1988)]. The preliminary experiment indicates that this perceptual accent shift is also observable in English; therefore, the effect must be taken into consideration in synthesizing English utterances. The present study investigates the case where the intonational H occurs on the utterance‐final syllable, e.g., “This is my nét.” In such a case, due to the exaggerated F0 lowering at utterance end, the drop rate after the peak may be so great that the perceptual shift can occur, i.e., the utterance may be perceived as “This is mý net.” Listening to synthesized utterances with varying drop rate on the final syllable, the subjects judged whether each utterance is more appropriate to responding to “What is it?” (inducing the accent on “net”) or “Is it her net?” (inducing the accent on “my”). The approximate value at which the shift occurs will be reported.

Declination in Toishan

Many have noted a tendency for fundamental frequency to gradually lower during the production of an utterance. This downtrend has sometimes been attributed to declination, a phonologically unmotivated lowering of F0 that is independent of tone. Recently, Johnson [Meeting Handbook, LSA, 21 (1985)] discovered data consistent with the view that declination exists in Cantonese. Like Cantonese, Toishan (another Yue dialect) exhibits little, if any tone sandhi. To see whether Toishan likewise exhibits declination, we had native speakers produce sentences containing high level tones, with varying amounts of material between the target high tones. The sentences are organized into passages so that, in addition to an examination of declination at the sentence level, an examination can be made at the discourse level as well (since it is possible that the degree of declination within a sentence depends on the position of the sentence within the discourse unit).

The influence of inter- and intra-speaker tempo on fundamental frequency and palatalization.

A study was conducted to examine the influence of speaking rate and emphatic stress on patterns of fundamental frequency (F0) and palatalization across word boundaries. Two groups of speakers, exhibiting characteristically fast versus slow speech, uttered preselected sentences at normal, fast, and slow rates. Acoustic analysis of F0 showed somewhat higher F0 peaks for characteristically fast speakers and for fast rates of speech. Emphatic stress was accompanied by a typical heightening of F0 on the emphasized word and by a lowering of F0 on a neighboring word. Palatalization across word boundaries was more frequent among characteristically fast speakers and at fast rates of speech. In addition, palatalization was more frequent in the absence of emphatic stress at the key site. The results for F0 were discussed in terms of the influence of vocal fold tension, while the results for palatalization were discussed in terms of the incompatibility of phonological processes of lengthening versus shortening as well as possible look-ahead restrictions.

Interaction between articulatory movements and vocal pitch control in Japanese word accent.

In order to investigate the interaction between articulatory and vocal pitch controls, EMGs of the cricothyroid (CT) and sternohyoid (SH) muscles were recorded during utterances of two-mora Japanese words with rising and falling accent types in the Tokyo dialects. The test words consisted of combinations of open and closed vowels. It was revealed that the changes in CT activity were consistently observed in correspondence with the F0 shift according to the accent types, while the SH muscle was observed to be active both for jaw opening and F0 lowering, with some subject-to-subject difference. It was also observed that the time lag between the decrease in CT activity and the F0 fall varied with a shift in the relative timing of SH activity in one subject, but not in the other. Articulatory jaw opening with SH activation was considered to positively contribute to F0 lowering, but the use of this mechanism appeared to be optional at least in the Tokyo accent.

A theoretical study of the effects of various laryngeal configurations on the acoustics of phonation.

Simulation of glottal volume flow and vocal fold tissue movement was accomplished by numerical solution of a time-dependent boundary value problem, in which nonuniform, orthotropic, linear, incompressible vocal fold tissue media were surrounded by irregularly shaped boundaries, which were either fixed or subject to aerodynamic stresses. Spatial nonuniformity of the tissues was of the layered type, including a mucosal layer, a ligamental layer, and muscular layers. Orthotropy was required to stabilized the vocal folds longitudinally and to accomodate large variations in muscular stress. Incompressibility and vertical motions at the golttis played an important role in producing and sustaining phonation. A nominal configuration for male fundamental speaking pitches was selected, and the regulation of fundamental frequency, intensity, average volume flow, and vocal efficiency was investigated in terms of variations around this nominal configuration. Parameters which were varied consisted of geometrical factors such as length, thickness, and depth, factors for shaping the glottis, as well as tissue elasticities, tissue viscosities, and subglottal pressure. Since nonlinear stress-strain properties were not included, subglottal pressure did not produce a pronounced effect upon fundamental frequency under these somewhat edealized conditions F0 rasing correlated strongly with increased tension in the ligament, and somewhat with increasing tension in the vocalis. F0 lowering correlated with increase in vocal fold length when the tensions were held constant, but not with increase in vocal fold thickness. Vocal intensity and efficiency are shown to have local maxima as the configurational parameters are varied one at a time. It appears that oral acoustic power output and vocal efficiency can be maximized by proper adjustments of longitudinal tension of nonmuscular (mucosal and ligamental) tissue layers in relation to muscular layers. Quantitative verification of the "body-cover" theory is therefore suggested, and several further implications with regard to control of the human larynx are considered.

Larynx height and word accent distinction in Japanese

The larynx by and large takes a higher position at the high voice fundamental frequency (F0) and moves downwards again with the lowering of F0. This paper focuses on the question whether variations in larynx height are correlated systematically with the different patterns of Japanese word accent. One male native speaker of standard Japanese pronounced phrases containing two‐mora test words of the form /VmV/ (where V stands for the vowel /a/ or /i/) while distinguishing among the three possible accent patterns for such sequences. Together with the speech sounds, lateral motion pictures of the subject's anterior neck were taken with a 16‐mm cinecamera. The differences in larynx height estimated from these pictures are discussed with respect to the physiological mechanisms for F0 control.

Physiological, correlates of intonation patterns.

An experiment has been performed to assess to what degree laryngeal muscle activity and subglottal air pressure affect the rate of vocal cord vibration in speech. Attention has been limited to those changes in the rate of vocal cord vibration that are associated with the physiological implementation of prosodic phenomena such as intonation and prominence. Subglottal air pressure was measured directly through a catheter inserted between the cricoid and thyroid cartilages. Using hooked-wire electrodes, the electromyographic activity was recorded in the right- and left-side cricothyroid muscle and in the sternohyoid, sternothyroid, and thyrohyoid muscles. The data were collected for one subject, a native speaker of Dutch. The results of the experiment show that, in this particular speaker, (1) cricothyroid muscle activity bears the most direct relationship to all major F0 changes, in that the contraction of this muscle raises F0 while its relaxation has an F0 lowering effect: (2) subglottal air pressure controls the gradually falling base line of the F0 contour and gives support to a rapid F0 drop if it occurs on the utterance-final syllable; and (3) the sternohyoid, sternothyroid, and thyrohyoid muscles have no direct effect on F0. Subject Classification: 70.20, 70.70.

Electromyographic Study of the Tones of Thai

In our desire to study the effects of laryngeal muscle activity on the rather complex use of fundamental frequency changes in a tone language, we have done an electromyographic analysis of the production of the tones of Thai. We chose the Thai language because its five phonemic tones differentiate words primarily by means of characteristic f0 contours. In the experiment, we asked four native speakers to say 45 short sentences. Included in this list were the five tones in nine contexts: the three long vowels /aa ii uu/ and the three labial consonants /b p ph/. Our findings indicate that for all four speakers, the activity of the cricothyroid muscle increases with the raising of f0, and the activity of the strap muscles—the sternohyoid, sternothyroid, and thyrohyoid muscles—increases with the lowering of f0. [This research was supported by the National Institute of Dental Research, NIH Grant.]